基于径向基神经网络的桥式起重机剩余寿命评估

EVALUATION FOR REMAINING LIFE OF BRIDGE CRANE BASED ON RADIAL BASIS NEURAL NETWORK

为降低起重机安全事故发生的概率,基于径向基神经网络提出一种快速计算起重机剩余寿命的方法。以某工厂的一台桥式起重机为例,根据实际参数建立Ansys有限元模型,通过现场实测数据对模型进行修正,进行静力学分析获取疲劳核算点位置。模拟起重机运行的典型工况,将小车位置及起吊载荷作为输入层,任意点的等效应力值作为输出层训练径向基神经网络模型,通过使用训练好的径向基神经网络模型来快速获取任意点的时间应力曲线,最后基于损伤容限断裂力学法进行剩余寿命评估。结果表明,通过利用径向基函数神经网络模型,可以实现对任意节点快速获取时间应力曲线,能够大大节省起重机现场实测的烦琐过程和大量投入,实现快速获取时间应力曲线从而计算出疲劳剩余寿命,完成桥式起重机疲劳剩余寿命估算,为起重机的长期安全使用和后期维修提供了可靠依据。

In order to reduce the probability of crane safety accidents, this paper proposes a method to quickly calculate the remaining life of the crane based on radial basis function(RBF)neural network. Taking a bridge crane in a factory as an example, an ANSYS finite element model is established based on actual parameters, and the model is modified through on-site measured data, and a static analysis is performed to obtain the location of the fatigue calculation point. Firstly, taking position of the trolley and the lifting load as input layer, the equivalent stress value at any point as output layer to stimulate the typical working conditions of crane operation. Secondly, to obtain time stress curve at any point quickly by using the well-trained RBF neural network model. Finally, to evaluate the residual life according to the damage tolerance fracture mechanics method. The results show that the time stress curve can be quickly obtained from any node by using the radial basis neural network model, which greatly decreased cumbersome process and save the cost in the crane site measurement, and realize the fast acquisition of the time stress curve to calculate the fatigue remaining life. Completing the estimation of the remaining fatigue life of the bridge crane provides a reliable basis for the long-term safe use and later maintenance of the crane.